What question did this study set out to answer?

This research aims to improve channel estimation in mmWave MIMO systems using a novel denoising framework.

January 22, 2026Open Access

HASwinNet: A Swin Transformer-Based Denoising Framework with Hybrid Attention for mmWave MIMO Systems

XHXi HanNorth China University of Technology HTHouya TuNorth China University of Technology JYJiaxi YingHong Kong University of Science and Technology

Key Points

This research aims to improve channel estimation in mmWave MIMO systems using a novel denoising framework.
Proposes HASwinNet, a deep learning framework for mmWave channel denoising.
Utilizes a hierarchical Swin Transformer encoder for structured representation learning.
Implements two branches: one for sparse token extraction and another for angular-domain refinement.
Employs discrete Fourier transform (DFT), squeeze-and-excitation (SE), and inverse DFT (IDFT) for feature enhancement.
Introduces an angular-domain perceptual loss during training to enhance spectral consistency.
Achieves significant improvements in normalized mean squared error (NMSE) and bit error rate (BER).
Outperforms CNN, LSTM, and U-Net baselines under various conditions.
Effectively exploits angular sparsity even with reduced pilot symbols, maintaining advantages over baselines.

Abstract

Millimeter-wave (mmWave) massive multiple-input, multiple-output (MIMO) systems are a cornerstone technology for integrated sensing and communication (ISAC) in sixth-generation (6G) mobile networks. These systems provide high-capacity backhaul while simultaneously enabling high-resolution environmental sensing. However, accurate channel estimation remains highly challenging due to intrinsic noise sensitivity and clustered sparse multipath structures. These challenges are particularly severe under limited pilot resources and low signal-to-noise ratio (SNR) conditions. To address these difficulties, this paper proposes HASwinNet, a deep learning (DL) framework designed for mmWave channel denoising. The framework integrates a hierarchical Swin Transformer encoder for structured representation learning. It further incorporates two complementary branches. The first branch performs sparse token extraction guided by angular-domain significance. The second branch focuses on angular-domain refinement by applying discrete Fourier transform (DFT), squeeze-and-excitation (SE), and inverse DFT (IDFT) operations. This generates a mask that highlights angularly coherent features. A decoder combines the outputs of both branches with a residual projection from the input to yield refined channel estimates. Additionally, we introduce an angular-domain perceptual loss during training. This enforces spectral consistency and preserves clustered multipath structures. Simulation results based on the Saleh–Valenzuela (S–V) channel model demonstrate that HASwinNet achieves significant improvements in normalized mean squared error (NMSE) and bit error rate (BER). It consistently outperforms convolutional neural network (CNN), long short-term memory (LSTM), and U-Net baselines. Furthermore, experiments with reduced pilot symbols confirm that HASwinNet effectively exploits angular sparsity. The model retains a consistent advantage over baselines even under pilot-limited conditions. These findings validate the scalability of HASwinNet for practical 6G mmWave backhaul applications. They also highlight its potential in ISAC scenarios where accurate channel recovery supports both communication and sensing.

اسأل الذكاء الاصطناعي

Bookmark

View Full Paper